1. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk ENGAGING STAKEHOLDERS WITH UNCERTAINTY ENGAGING STAKEHOLDERS WITH UNCERTAINTY Hazel Faulkner & Simon McCarthy ?

2. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk FHRC research for FRMRC1/2 INTERVIEWS with Environment Agency professionals Environment Agency professionals Insurers Insurers Floodplain planners Floodplain planners  How can FRM optimise risk and uncertainty communications at the professional interface?  Why the reluctance to use uncertainty tools....barriers?

3. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk BARRIERS TO EFFECTIVE COMMUNICATION OF UNCERTAINTY BARRIERS?

4. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk Stakeholders in information pathways As webs of influence vary, agendas vary, information needs vary, risk communication strategies & appropriate tools will be totally different MANY SCIENTIFIC,PROFESSIONAL AND LAY STAKEHOLDERS stakeholders at sources of information ‘science’ professionals Information receptors, category I floodplain stakeholders Information receptors, Category II stakeholders outside floodplain BARRIERS?

5. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk  The implication of this is that scientific uncertainty is an relatively unwelcome part of the risk message they are charged with translating for the public PROFESSIONALS experience BINARY DECISION UNCERTAINTY (Decision rule uncertainty) BARRIERS? SCIENTIFIC UNCERTAINTY Uncertainty in the science of flood forecasting and runoff prediction models are largely associated with their assumptions, structure, and boundary conditions, and confidence in validation procedures given uncertainties about climatic and societal futures SCIENTIFIC AND DECISION MAKERS DIFFER IN THEIR UNDERSTANDING OF THE SOURCES OF UNCERTAINTY

6. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk THE LANGUAGE OF SCIENCE IS OPAQUE TO THE NON-SCIENTIFIC PROFESSIONAL BARRIERS?  If originally formulated in Bayesian terms, the language may be too opaque for translation to be effective – do professionals have sufficient statistical familiarity? Bayesian statistics ? Bayesian statistics ? Prior probability distributions? Prior probability distributions? fuzzy set methods ? fuzzy set methods ? info-gap methods ? info-gap methods ? NUSAP? NUSAP? match tool to communication interface match tool to communication interface  The implication of this is that the language used to communicate the uncertainty must match the needs and agenda of the agencies involved in the communication being undertaken

7. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk Stakeholders in information pathways Bayesian uncertainty tools – GLUE stakeholder consultation ’traffic lights’ webpages/leaflets Rt = H x V 1;100/1;1000 fuzzy edged Phone warnings Newspaper/TV /twitter LANGUAGE AND TOOLS TO DESCRIBE UNCERTAINTY VARY stakeholders at sources of information ‘science’ professionals Information receptors, category I floodplain stakeholders Information receptors, Category II stakeholders outside floodplain THE WAY FORWARD: TRANSLATIONAL DISCOURSES?

8. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk Our research (Faulkner et al 2007;McCarthy et al. 2009), both the interviews undertaken with professionals and from the ‘experiment’ undertaken at the co-location workshop in Exeter, found that :  The power of VISUALISATIONS AND ANIMATIONS in realising the uncertainty estimates was potentially great;  The professionals questioned initially struggled to comprehend scientifically defined flood forecast uncertainties (probabilistic and/or ensemble forecasts) without FURTHER TRANSLATION OF THE SCIENCE. When this was available as perhaps a ‘translational discourse’, the preparedness to embrace a more sophisticated expression of the model’s uncertainties was welcomed.  Better DECISION-SUPPORT TOOLS are needed. FINDINGS FINDINGS

9. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk Decision Tree and Wiki Pages at http://www.floodrisknet.org. uk/methods/ http://www.floodrisknet.org. uk/methods/ The WIKI Decision-support tool

10. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk The position of the translational discourse straddling domain  Alternative extended communication structures will be needed (a shared domain)  New interdisciplinary languages and professional challenges may need to exist;  A shared ‘‘virtual institution’’ (alternatively, the arena may be envisaged as having physical construct);

11. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk IMPROVED VISUALISATION OF UNCERTAINTY THE WAY FORWARD: V ISUALISATIONS

12. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk ‘end-user’ tailored From this... To this.... THE WAY FORWARD: V ISUALISATIONS

13. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk  Although uncertainty modelling is still not routine,1 and 2D inundation simulations with uncertainty layers ‘underneath’ are a possibility visually clear visually clear  Ensemble rainfall forecasts ( e.g. European Centre for medium range Weather Forecasts) THE WAY FORWARD: V ISUALISATIONS interrogatable

14. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk SHAPING A ‘TRANSLATIONAL DISCOURSE’ A ‘TRANSLATIONAL DISCOURSE’

15. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk A GUIDANCE MANUAL how to involve all stakeholders, especially professionals at local level is needed – professional input here important; MORE DECISION SUPPORT TOOLS NEEDED...... BRAINSTORMING Brainstorming sessions involving professionals/scientists and practitioners A ‘TRANSLATIONAL DISCOURSE’

16. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk references : flooding & uncertainty Faulkner, Parker, D., Green, C., and Beven, K. 2007. Developing a Translational Discourse to Communicate Uncertainty in Flood Risk between Science and the Practitioner. Ambio 36(7), 692-703, December 2007 Lane SN, Landström C, Whatmore SJ: 2011 Imagining flood futures: risk assessment and management in practice. Philos Transact A Math Phys Eng Sci; 13;369(1942):1784-806 Romanowicz, R and Beven, K J, 1998, Dynamic real-time prediction of flood inundation probabilities, Hydrol. Sci. J., 43(2), 181-196. Pappenberger, F., Beven, K.J., Hunter N., Gouweleeuw, B., Bates, P., de Roo, A., Thielen, J., 2005, Cascading model uncertainty from medium range weather forecasts (10 days) through a rainfall-runoff model to flood inundation predictions within the European Flood Forecasting System (EFFS). Hydrology and Earth System Science, 9(4),381-393 Note: Klir has taken forward the debate. In the UK scientific community these ideas are referred to in papers by Hall Pappenburger et al, and Pappenburger and Beven. At the Department for Environment, Food, and Rural Affairs (Defra and EA ) several reports are available.

17. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk Acknowledgement The research reported in this presentation was conducted as part of the Flood Risk Management Research Consortium with support from the: – Engineering and Physical Sciences Research Council – Department of Environment, Food and Rural Affairs/Environment Agency Joint Research Programme – United Kingdom Water Industry Research – Office of Public Works Dublin – Northern Ireland Rivers Agency Data were provided by the EA and the Ordnance Survey.

18. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk The’ translational discourse’ is a mechanism for the development of tools to match needs. COUCHED in : Mutually exchanged and agreed languages and professional context; EMBRACES: A decision about the modelling strategy to be adopted for a particular type of application, any decision about which uncertainties should be modelled should also be made by joint exchanges within the translational discourse. A ‘TRANSLATIONAL DISCOURSE’

19. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk  In the Netherlands, a broader understanding of the errors and biases that permeate human perception of risk information has emerged, and new ‘‘postnormal’’ communication tools have been debated (de Sluijs, et al), and these may be of value in assessing model uncertainty  Wiki pages, decision-support trees developed by FRMRC2 group at Lancaster  Good Practice guide has been developed.

20. Funders: EPSRC Grant: EP/FP202511/1 www.floodrisk.org.uk  Challenge of USING MODELS IN SERIES When models are linked in series, uncertainties become cascaded in ways that can be only be partially constrained. The actual uncertainty may considerably exceed the reported uncertainty BARRIERS TO EFFECTIVE COMMUNICATION OF RISK AND UNCERTAiNTY cont.. BARRIERS TO EFFECTIVE COMMUNICATION OF RISK AND UNCERTAINTY at THE PROFESSIONAL INTERFACE cont..